Optimisation - Is it a fallacy? - page 3

[[Deleted]]

That is exactly what I meant. MA-crosses don't work, because if they did work, there should have been some correlation between results from one month to the other. My idea is that some indicators can work, but only in some kind of market. If the market is volatile, breakout-theories seem to work. However when the market isn't volatile, the same indicators/theories don't work. So you first have to know which kind of market you are in. And when you know that, pick the right system for that market.

[[Deleted]]

Ah, but how do you know what kind of market you are in? By using something like the Directional Movement System? Again, the problem with all these indicators is lag.. By the time the DMS has worked out you're in a trend, the trend is nearly finished again. So while you may still make small profits on those trades, the false starts, whipsaws etc that you inevitably end up with, more than wipe out your profits.

I'm getting to a point now where I'm inclined to agree with Craig. I'm starting to think that the only way to CONSISTANTLY trade sucessfully is to trade the long term trends, using fundimentals as a basis and maybe a bit of TA (i.e. literally only long term breakouts, trendlines and maybe the odd chart pattern) to enter and exit trades. Although having said that, I can imagine that news trading is probably not a bad way to go although again, show me someone who is CONSISTANTLY profitable.

But, isn't it interesting to note that the hedge funds (i.e. groups who employ the brightest of the bright) trade long term? Now why would that be?